Advances in cancer detection and treatment have shifted cancer from a short-term, fatal disease, to a chronic condition necessitating a focus on longer-term outcomes. One-third to one-half of cancer patients experience lasting psychological distress (i.e. depression and anxiety). Distress predicts increased health care utilization at an excess annual cost of $2,000-3,000 per patient. Previous studies however, have severe methodologic limitations, including study designs limited to cancer patient follow-up, an emphasis on younger adult cancer survivors, and narrow stress definitions that do not incorporate a comprehensive stress assessment that includes measures of both positive and negative affect. The broader extent to which an adverse stress response is associated with health care utilization and health outcomes in the vulnerable older adult cancer patient population is unknown. Further, a clinical tool to help identify at-risk patients at the time of diagnosis does not exist. This is a critical limitation, as fewer than 10 percent of patients are actively referred for supportive services, suggesting that cancer-related stress is inadequately addressed in the clinic setting. The aims of this project are to: (1) assess the magnitude of increase in psychological stress symptoms pre- to post- cancer diagnosis;(2) determine whether patients at risk of a high or sustained stress response can be identified at the time of a cancer diagnosis;and (3) assess the effect that increased stress symptoms have on patient health care utilization and health outcomes. Data from the Wisconsin Longitudinal Study (WLS), a long- term study of a one-third random sample of men and women who graduated from Wisconsin high schools in 1957 and a randomly selected sibling (n=7,113), are used to accomplish these aims. Eleven percent of WLS respondents are cancer survivors. The analytic plan estimates causal effects by capitalizing on the WLS study design (longitudinal data, availability of a non-cancer control group, and paired sibling respondents) and existing linkages to the Wisconsin Tumor Registry (1979-2006) and Medicare databases (2004-06). Residualized stress difference scores will be used as an outcome variable in multivariate regressions to assess the magnitude of the cancer stress response and the individual and situational characteristics associated with an elevated or sustained response. These difference scores will then be interacted with cancer status to predict health care utilization (clinic/hospital and preventive care) and health outcomes. The proposed research directly relates to the NCI's mission by examining psychological influences on cancer and cancer-related behavior and by initiating development of a tool to identify at-risk patients of an elevated and sustained stress response, thereby improving cancer care delivery and patient adjustment to the effects of cancer and its treatment. The results from our investigation will have important implications for health care, allowing for the effective and efficient targeting of tailored programs to those at highest risk, thereby improving patient quality of life while reducing costly health care use. PUBLIC HEALTH RELEVANCE: Project Narrative Identifying older cancer patients at risk of an elevated stress response at the time of diagnosis is critical in order to maximize high quality care;however a clinical tool for identifying older patients at risk of a high stress response does not exist. We propose to estimate the nature, magnitude, and duration of the stress response pre- to post-cancer diagnosis and determine individual and situational factors associated with an elevated response, while simultaneously assessing its health care utilization and health outcome consequences. The results from our investigation will have important implications for patient care, including reducing patient suffering and improving quality of life while reducing excess health care utilization and its commensurate costs.